Skip to content

Implement Unet and its variants from scratch using PyTorch. Train those models and compare the results.

Notifications You must be signed in to change notification settings

HJoonKwon/unet-torch-mps

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

93 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

unet-torch-mps

Models

Installation

Install PyTorch that corresponds with your device.

$ conda install pytorch torchvision torchaudio pytorch-cuda=12.1 -c pytorch -c nvidia

Install dependencies

if something's missing, please install them with pip.

$ pip install -r requirements.txt

RUN

$ cd scripts && bash download.sh

Go to unet-torch-mps/python and run

$ python train.py --help

usage: train.py [-h] [-d D] [-lr LR] [-e E] [-c C] [-b B] [-ci CI] [-sanity_check] [-ckpt CKPT]

options:
  -h, --help     show this help message and exit
  -d D           dataset directory
  -lr LR         learning rate
  -e E           number of epochs
  -c C           number of classes
  -b B           batch size
  -ci CI         ckpt interval
  -sanity_check  sanity check mode. use validation data for training
  -ckpt CKPT     ckpt path

Sample Result

https://wandb.ai/joon/unet-pytorch

image

TODO

  • implement scheduler
  • implement attention unet
  • optimize gpu utilization
  • make docker file

About

Implement Unet and its variants from scratch using PyTorch. Train those models and compare the results.

Topics

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published